Dendreon Corp fka DNDNQ

[rancherho]

Re: 9902b event target and timing
p3analyze:

1. You have abundantly demonstrated your knowledge of biostatistics in the past (the musing about the 4 Prentice criteria for surrogacy being a great example), and I am defintely no statistician, so what follows is a question regarding your analysis.

2. The statistical significance of 9902b is to be determined by the Cox regression analysis where we know the pooled 225 patients 9901/9902a p value was 0.0006 vs. the Kaplan Meier p value of 0.011, in spite of a placebo crossover rate somewhat in excess of 70% It would seem to me that if one assumes the same treatment effect and event rate for 9902b as for the pooled 9901 and 9902a enrollees, the target p value threshold of 0.05 would be reached with substantially fewer events than if a Kaplan Meier derived p value was needed. Would this be true?

3. Although not exactly on point for 9902b since any trial unblinding will likely only be precipitated by reaching a prespecified number of events event, the following citation points to a somewhat comparable case where trials were stopped short of full planned enrollment due to unexpectedly high treatment effects, a caveat being that statistical data can have spikes during the trial period and stopping a trial short of full enrollment could be precipitated by such a spike.Randomized Trials Stopped Early for Benefit A Systematic Review http://jama.ama-assn.org/cgi/content/full/294/17/2203

4. A Cox regression analysis corrects, of course, for certain prespecified variables that its calculations determine to be statistically signficant at a set rate (p=0.05 in 9901/9902a, p=0.010 in the Taxotere TAX327 trial). In the 9901/9902a trials, treatment effect was the most statistically significant. If one assumes that the placebo crossover for 9902b might increase from somewhat over 70% to 80% in 9902b due to an enlargement of the acceptable patient enrollment criteria, there could be 100 true placebos in the final 9902b trial. Both ITT and crossover placebo patients would receive Provenge, and thus have data to calculate their cumulative CD54 cell counts and cumulative CD54 upregulation. There should be some analysis that could be done, admittedly retrospective, that would censor non AIPC deaths, include the crossover patients in the ITT group and then calculate and plot a sensitivity analysis of the impact of the spectrum of the new ITT group's cumulative CD54 cell count and cumulative CD54 upregulation on survival vs. true placebos where the underlying Cox regression analysis has preumably corrected for the relative health of all enrollees. It seems to me that this would suggest where increasing initial Provenge dosing might be appropriate. Subsequent booster dosing probably would require some ongoing immunity monitoring, including T cell proliferation, memory T cell and Treg levels.
Just a thought.

Good luck to you and to all DNDN longs.